1/7 Serosurveys! We scientists are quibbling about specificity, but what do these study results mean for the general public?

My take on what serosurveys can and cannot tell us.

2/7 Emerging data are consistently showing that seroprevalence in most areas is low. Only in hard-hit areas are we seeing prevalence above single digits. This means we are a ways off from herd immunity levels (60%+).

3/7 It’s clear that there are many more infections than laboratory confirmed cases. I guess 10-20x, and others feel similar, though this may change over time. We expected under-reporting because of incomplete testing and because mild and asymptomatic infections are missed.

4/7 The infection fatality ratio (IFR) is much lower than the crude case fatality ratio of 5% in the US. IFR is likely <1%. But there is still too much uncertainty in both the numerator (underreported deaths) and denominator (infections) to establish an exact value.

5/7 Even if the IFR is 1 in 1000, the sheer numbers of people who are susceptible make this virus devastating. The virus spreads easily and quickly. We need only look at the numbers hospitalized and dead in hard-hit areas to see its impact.

6/7 What can serosurveys tell us? They can help us identify high risk populations most likely to be infected and most likely to die given infection. We are seeing important differences across race/ethnicities, ages, and underlying conditions.

7/7 What can serosurveys not tell us? Whether we are ready to re-open our economy. This will depend on having a robust system to test, trace, and isolate infections. Without this, we are no better off than we were before, and would expect the virus to quickly resurge. END